The web link selection problem is to select a small subset of web links from a large web link pool, and to place the selected links on a web page that can only accommodate a limited number of links, e.g., advertisements, recommendations, or news feeds. Despite the long concerned click-through rate which reflects the attractiveness of the link itself, the revenue can only be obtained from user actions after clicks, e.g., purchasing after being directed to the product pages by recommendation links. Thus, the web links have an intrinsic \emph{multi-level feedback structure}. With this observation, we consider the context-free web link selection problem, where the objective is to maximize revenue while ensuring that the attractiveness is no less than a preset threshold. The key challenge of the problem is that each link's multi-level feedbacks are stochastic, and unobservable unless the link is selected. We model this problem with a constrained stochastic multi-armed bandit formulation, and design an efficient link selection algorithm, called Constrained Upper Confidence Bound algorithm (\textbf{Con-UCB}), and prove $O(\sqrt{T\ln T})$ bounds on both the regret and the violation of the attractiveness constraint. We conduct extensive experiments on three real-world datasets, and show that \textbf{Con-UCB} outperforms state-of-the-art context-free bandit algorithms concerning the multi-level feedback structure.

Toronto's status as a burgeoning tech and artificial intelligence hub recently received a huge stamp of validation. The community made the final list of 20 cities in the running for Amazon's HQ2, earning that distinction without offering Amazon any tax breaks or financial incentives. Being a tech torch bearer is nothing new for Canada. In the early 2000s, BlackBerry and Nortel held sizable market shares in smartphones and telecom, respectively. Not long after that, though, an inability to gain a competitive advantage closed the door on each brand's chance to dominate the marketplace.

School violence has increased over the past ten years. This study evaluated students using a more standard and sensitive method to help identify students who are at high risk for school violence. Participants (ages 12–18) were active students in 74 traditional schools (i.e. Collateral information was gathered from guardians before participants were evaluated. School risk evaluations were performed with each participant, and audio recordings from the evaluations were later transcribed and manually annotated.

A pilot study indicates that artificial intelligence may be useful in predicting which students are at higher risk of perpetrating school violence. The researchers found that machine learning - the science of getting computers to learn over time without human intervention - is as accurate as a team of child and adolescent psychiatrists, including a forensic psychiatrist, in determining risk for school violence. "Previous violent behavior, impulsivity, school problems and negative attitudes were correlated with risk to others," says Drew Barzman, MD, a child forensic psychiatrist at Cincinnati Children's Hospital Medical Center and lead author of the study. "Our risk assessments were focused on predicting any type of physical aggression at school. We did not gather outcome data to assess whether machine learning could actually help prevent school violence. That is our next goal."

In this #EdOnEdtech following video, Dr. ET discusses the past, present, and future of AI as told by those who are shaping the definition of personalized learning. Discover the present perspectives on personalized learning from leaders in education and technology including Mark Zuckerberg, Bill Gates, McGraw Hill's CEO - David Levin, James Shelton, and Betsy DeVos. Check out the future of artificial intelligence and personalized learning from those in the process of funding it, such as Mark Zuckerberg and Bill Gates. We are excited and honored that NUITEQ's Dr. ET has interviewed Richard Wells in the EdTech Lounge. Richard is an EdTech influencer who founded EduWells, a top 10 education blog.

Artificial intelligence (AI) has been changing our lives for decades, but today its presence is bigger than ever before. Sometimes, we don't even realize it when a new AI-powered system, tool, or product appears and outperforms us, humans. The benefits of AI and automation for HR and the workforce don't come instantly, however. It's a journey and one can see the short-term benefits of this journey in automation, the medium-term benefits in augmentation and finally the long-term benefits in the amplification of human activities or tasks. Let's take a look at the various effects of AI and automation on HR and the workforce in more detail.

On Wednesday, a group of tech industry and academic leaders released a new benchmark to measure the speed of machine learning software and hardware and accelerate improvements in system performance. The benchmark--called MLPerf--was created by a number of tech companies including AMD, Baidu, Google, and Intel, as well as researchers from educational institutions including Harvard, Stanford, and the University of California Berkeley. MLPerf measures speed based on the time it takes to train deep neural networks to perform tasks such as recognizing objects, translating languages, and playing the game of Go. SEE: IT leader's guide to the future of artificial intelligence (Tech Pro Research) As machine learning and artificial intelligence (AI) efforts grow across the enterprise, systems need to evolve quickly to meet demands, Andrew Ng, founder and CEO of Landing.AI, said in a press release. "AI is transforming multiple industries, but for it to reach its full potential, we still need faster hardware and software," Ng said in the release.

Here I offer a few off-the-beaten-path interesting problems that you won't find in textbooks, data science camps, or in college classes. These problems range from applied maths, to statistics and computer science, and are aimed at getting the novice interested in a few core subjects that most data scientists master. The problems are described in simple English and don't require math / stats / probability knowledge beyond high school level. My goal is to attract people interested in data science, but who are somewhat concerned by the depth and volume of (in my opinion) unnecessary mathematics included in many curricula. I believe that successful data science can be engineered and deployed by scientists coming from other disciplines, who do not necessarily have a deep analytical background yet are familiar with data.

The explosion of AI capabilities and other emerging technologies is clearly transforming the practice of law. Can these technologies also be leveraged to prepare students for an evolving job market? Working closely with our partners at Thomson Reuters, we at Above the Law have been exploring the impact of AI and other technologies on law schools. We now invite you to explore Cognifying Legal Education, the first in a four-part, multimedia exploration of how artificial intelligence and similar innovations are reshaping the legal profession: Law2020.

We describe a novel family of models of multi- layer feedforward neural networks in which the activation functions are encoded via penalties in the training problem. Our approach is based on representing a non-decreasing activation function as the argmin of an appropriate convex optimiza- tion problem. The new framework allows for algo- rithms such as block-coordinate descent methods to be applied, in which each step is composed of a simple (no hidden layer) supervised learning problem that is parallelizable across data points and/or layers. Experiments indicate that the pro- posed models provide excellent initial guesses for weights for standard neural networks. In addi- tion, the model provides avenues for interesting extensions, such as robustness against noisy in- puts and optimizing over parameters in activation functions.